Current inductive learning algorithms have difficulties handling attributes with numerical values. This paper presents RULES-F, a new fuzzy inductive learning algorithm in the RULES family, which integrates the capabilities and performance of a good inductive learning algorithm for classification applications with the ability to create accurate and compact fuzzy models for the generation of numerical outputs. The performance of RULES-F in two simulated control applications involving numerical output parameters is demonstrated and compared with that of the well-known fuzzy rule induction algorithm by Wang and Mendel.
QuinlanJ. R.Learning with continuous classes. Proceedings of the 5th Australian Joint Conference on Artificial intelligence, World Scientific, Singapore, 1992, pp. 343–348.
2.
BreimanL.FriedmanJ. H.OlshenR. A.StoneC. J.Classification and regression trees, 1984 (Wadsworth amp; Brooks/Cole Advanced Books amp; Software, Belmont, CA, USA).
3.
MorimotoY.IshiiH.MorishitaS.Efficient construction of regression trees with range and region splitting. Proceedings of the 23rd Very Large Data Bases Conference, Athens, Greece, 1997, pp. 166–175.
4.
KaralicA.CestnikB.The Bayesian approach to tree-structured regression. Proceedings of Information Technology Interfaces 91, Cavtat, Yugoslavia, 1991, pp. 155–160.
5.
BruhaI.BerkaP.Continuous classes in rule induction: Empirical comparison of two approaches. 3rd International Workshop on Artificial intelligence techniques, Brno, Czech Republic, 1996, pp. 163–164.
6.
KaralicA.Employing linear regression in regression tree leaves. Proceedings of European Conference on Artificial intelligence 92, Vienna, Austria, 1992, pp. 440–441.
7.
SebagM.SchoenauerM.Inductive learning of membership functions and fuzzy rules. Uncertainty modeling: Theory and applications. Machine intelligence and pattern recognition series, 1994, pp. 275–283 (North Holland, Netherlands).
8.
NozakiK.IshibuchiH.A simple but powerful heuristic method for generating fuzzy rules from numerical data. Fuzzy Sets Syst., 1997, 86, 251–270, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
9.
PhamD. T.BigotS.DimovS. S.RULES-5: A rule induction algorithm for classification problems involving continuous attributes. Proc. Instn Mech. Engrs. Part C: J. Mechanical Engineering Science, 2004, 217, 1273–1286.
10.
WangL. X.MendelJ. M.Generating fuzzy rules from numerical data, with applications. Technical Report TR USC-SIPI #169, Signal and Image Processing Institute, University of Southern California, CA, USA, 1991.
11.
DelgadoM.GonzalezA.An inductive learning procedure to identify fuzzy systems. Fuzzy Sets Syst., 1993, 55, 121–132, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
12.
HongT. P.LeeC. Y.Induction of fuzzy rules and membership functions from training examples. Fuzzy Sets Syst., 1996, 84, 33–47, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
13.
HongT. P.LeeC. Y.Effect of merging order on performance of fuzzy induction. Intell. Data Anal., 1999, 3 (2), 139–151.
14.
HongT. P.ChenJ. B.Processing individual fuzzy attributes for fuzzy rule induction. Fuzzy Sets Syst., 2000, 112, 127–140, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
15.
ShannJ. J.FuH. C.A fuzzy neural network for rule acquiring on fuzzy control systems. Fuzzy Sets Syst., 1995, 71, 345–357, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
16.
JengB.JengY. M.LiangT. P.FILM: A fuzzy inductive learning method for automated knowledge acquisition. Deci. Support Syst., 1997, 21 (2), 61–74.
17.
QuinlanJ. R.Induction of decision trees. Mach. Learn., 1986, 1, 81–106.
18.
CastroJ. L.Castro-SchezJ. J.ZuritaJ. M.Use of a fuzzy machine learning technique in the knowledge acquisition process. Fuzzy Sets Syst., 2001, 123, 307–320, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
19.
WangC. H.LiuJ. F.HongT. P.TsengS. S.A fuzzy inductive learning strategy for modular rules. Fuzzy Sets Syst., 1999, 103, 91–105, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
20.
RaviV.ZimmermannH. J.ReddyP. J.Fuzzy rule base generation for classification and its minimization via modified threshold accepting. Fuzzy Sets Syst., 2001, 120, 271–279, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
21.
WangX. Z.WangY. D.XuX. F.LingW. D.YeungD. S.A new approach to fuzzy rule generation: Fuzzy extension matrix. Fuzzy Sets Syst., 2001, 123, 291–306, available from (accessed 2 August 2004). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
22.
RondeauL.RuelasR.LevratL.LamotteM.A defuzzification method respecting the fuzzification. Fuzzy Sets Syst., 1997, 86, 311–320, available from (accessed 3 October 2002). https://www-elsevier-com.web.bisu.edu.cn/locate/fss.
23.
PhamD. T.BigotS.DimovS. S.RULES-5: A rule merging technique for handling noise in inductive learning. Proc. Instn Mech. Engrs. Part C: J. Mechanical Engineering Science, 2003, 217, 1273–1286.
24.
ArmstrongB.KhatibO.The explicit dynamic model and inertial parameters of the PUMA 560 Robot arm. International Conference on Robotics and automation. San Francisco. CA. USA, 1986, Vol. 1, pp. 510–518.
25.
CraigJ. J.Introduction to robotics: Mechanics and control, 3rd edition, 2004 (Prentice Hall, Englewood Cliffs, NJ).
